Abuse Level Multiclass Text Classification Model

📆 CS5950 Machine Learning Course

• Explored and compared different machine learning algorithms and techniques, including word embeddings, recurrent neural networks, and support vector machines, to identify the best approach for detecting abusive speech with the ConvAbuse dataset
• Utilized validation techniques such as Holdout Validation. K-fold Cross-Validation, and Leave One Out Cross-Validation to improve the model resulting in an accuracy of > 80% on the test dataset